GIScience Logo
HeiGIT Logo
3DGeo Logo

GIScience Blog

Tag: deep learning

  • ISPRS IJGI highlights our work on deep learning of Street Art from VGI and Street View Images

    We are pleased to share that because of the response to our work, ISPRS IJGI selected our paper on Detecting Graffiti with Street View Images and Deep Learning to be highlighted as a title story through some graphics on the journals main page. Novack T, Vorbeck L, Lorei H, Zipf A. (2020): Towards Detecting Building Facades…

  • Detecting OSM Building Facades with Graffiti Artwork Based on Street View Images and Social Media using Deep Learning

    As a recognized type of art, graffiti is a cultural asset and an important aspect of a city’s aesthetics. As such, graffiti is associated with social and commercial vibrancy and is known to attract tourists. However, positional uncertainty and incompleteness are current issues of open geo-datasets containing graffiti data. In a newly published paper, we…

  • Invited talk at the Spatial Data Science Symposium 2019, Santa Barbara

    The Center for Spatial Studies, Department of Geography at the University of California, Santa Barbara is hosting the Spatial Data Science Symposium 2019 this coming week with the title “Setting the Spatial Data Science Agenda” Over 40 selected participants will gather to discuss the future of Spatial Data Science at this expert meeting. Instead of…

  • New DFG project: IdealVGI – Deep Learning with OSM

    Recently a new DFG project proposal was accepted to the GIScience Research Group Heidelberg within the DFG priority programme VisVGI (Volunteered Geographic Information: Interpretation, Visualisation and Social Computing” [SPP 1894]). It is joint collaboration project together with Prof. Begüm Demir from TU Berlin. IDEAL-VGI: Information Discovery from Big Earth Observation Data Archives by Learning from…

  • Classification of 3D ALS Point Clouds using End-To-End Deep Learning

    In a new publication, we show how deep neural networks can be used in an end-to-end manner for the classification of 3D point clouds from airborne laser scan data. The research, based on the award-winning diploma thesis of Lukas Winiwarter at TU Wien, has now been published in “PFG – Photogrammetrie, Fernerkundung, Geoinformation“, the Journal…

  • Mapping Human Settlements with Higher Accuracy and Less Volunteer Efforts by Combining Crowdsourcing and Deep Learning

    Our new paper on Machine Learning and Humanitarian Mapping Nowadays, Machine Learning and Deep Learning approaches are steadily gaining popularity within the humanitarian (mapping) community. New tools such as the ML Enabler or the rapId editor might change the way crowdsourced data is produced in the future. Hence, at the Heidelberg Institute for Geoinformation Technology…

  • Estimating OpenStreetMap Missing Built-up Areas using Pre-trained Deep Neural Networks

    Recently a new paper about Estimating OpenStreetMap Missing Built-up Areas using Pre-trained Deep Neural Networks (DNNs) has been presented at the AGILE GIScience conference 2019 in Cyprus. Although built-up areas cover only a small proportion of the earth’s surface, these areas are closely tied to most of the world’s population and the economic output, which makes…

  • 3DGeo at the Geospatial Week 2019

    This week, the 3DGeo participated in the ISPRS Geospatial Week 2019 with two presentations among the sessions of the Laser Scanning Workshop with many interesting talks and poster. Presentations were given by Ashutosh Kumar in the Machine Learning Session and Katharina Anders in the Change Detection Session. Highlight: The work by Ashutosh Kumar on feature…

  • Paper on Analysis of Feature Relevance in Deep Learning for 3D Point Cloud Classification

    A paper investigating the relevance of (pre-calculated) features for 3D point cloud classification using deep learning was just published in the ISPRS Annals of Photogrammetry and Remote Sensing. The study presents a non-end-to-end deep learning classifier for 3D point clouds using multiple sets of input features and compares it with an implementation of the state-of-the-art…

  • PhD Colloquium “Deep Learning in Photogrammetry, Remote Sensing and Geospatial Information Processing”

    On 14th and 15th May, our 3DGeo group members Bernhard Höfle and Lukas Winiwarter were co-organizing and participating in the 4th colloquium for PhD students working on the topic of Deep Learning and its applications in Photogrammetry, Remote Sensing and Geoinformation Processing of the Deutsche Geodätische Kommission (DGK) and the Deutsche Gesellschaft für Photogrammetrie und…

  • Put the world’s most vulnerable people on the map with MapSwipe

    Humanitarian organizations can’t help people if they can’t find them. This was the simple reason to create MapSwipe back in 2016 and it is still as pressing as in the very beginning. In the last 2,5 years volunteers have contributed more than 18,000,000 results, which help humanitarian organizations to create maps of human settlements and…

  • Hyperspectral and LiDAR Fusion Using Deep Three-Stream Convolutional Neural Networks

    Our feature paper “Hyperspectral and LiDAR Fusion Using Deep Three-Stream Convolutional Neural Networks” is now published online. Recently, convolutional neural networks (CNN) have been intensively investigated for the classification of remote sensing data by extracting invariant and abstract features suitable for classification. In this paper, a novel framework is proposed for the fusion of hyperspectral…